The invention relates generally to predicting a cognitive state of a user, and more particularly to building synthetic events from measured data sources, such as sensors, and processing these synthetic events to denote a probabilistic inference.
The amount of measure data is growing at an exponential rate. Many times this results in too much data and too many kinds of data. The data is often unlabeled and uncatalogued. Moreover, in entities with different levels, such as a corporation, there may specific rules or procedures for each level, e.g. department rules, functional area rules, division rules, and corporate wide rules. Many times every level of every entity has rules applicable to their data. Rules are often conflict one with another. The application of rules to data become combinatoric explosive.
The is currently no efficient and effective to go from sets of data and at various levels and combine them into labelled, repeatable and definable events. Next take those events and convert them into a probabilistic inference.